Allow untracked .ttl files in --prov-from-git

CHANGELOG.md

@@ -1,5 +1,11 @@
 # Change log
 
+## Release 1.0.2 (2026-02-09)
+
+Fixes:
+
+- Allow untracked .ttl files in `--prov-from-git`. Skip them with an info log instead of raising an error. [#339](https://github.com/nfdi4cat/voc4cat-tool/issues/339)
+
 ## Release 1.0.1 (2026-01-06)
 
 Fixes:

src/voc4cat/transform.py

@@ -166,8 +166,7 @@ def add_prov_from_git(
         source_dir: Directory to look up git history from (if different from vocab_dir).
             Used when files have been copied to a new location.
 
-    Raises:
-        Voc4catError: If any .ttl file is not tracked in git.
+    Untracked .ttl files are skipped with an informational log message.
     """
     repo_dir = Path(repo_dir) if repo_dir else Path.cwd()
     vocab_dir = Path(vocab_dir)
@@ -182,21 +181,6 @@ def add_prov_from_git(
     # Get git info from the source directory (or vocab_dir if no source specified)
     git_info = get_directory_git_info(git_lookup_dir, repo_dir)
 
-    # Check that all .ttl files are tracked in git (by relative path)
-    for ttl_file in ttl_files:
-        # Preserve subdirectory structure when looking up in source directory
-        rel_to_vocab = ttl_file.relative_to(vocab_dir)
-        source_file = git_lookup_dir / rel_to_vocab
-        try:
-            rel_path = source_file.relative_to(repo_dir)
-        except ValueError:
-            rel_path = source_file
-        # Normalize path separators (git uses forward slashes)
-        rel_path_str = str(rel_path).replace("\\", "/")
-        if rel_path_str not in git_info:
-            msg = f'File "{ttl_file}" is not tracked in git. Cannot add provenance.'
-            raise Voc4catError(msg)
-
     # Process each .ttl file
     for ttl_file in ttl_files:
         # Preserve subdirectory structure when looking up in source directory
@@ -207,6 +191,11 @@ def add_prov_from_git(
         except ValueError:
             rel_path = source_file
         rel_path_str = str(rel_path).replace("\\", "/")
+        if rel_path_str not in git_info:
+            logger.info(
+                'File "%s" is not tracked in git. Skipping provenance.', ttl_file
+            )
+            continue
         info = git_info[rel_path_str]
 
         # Parse the RDF graph

tests/test_transform.py

@@ -222,8 +222,8 @@ def test_prov_from_git_adds_dates(git_repo_with_split_files, monkeypatch, caplog
     assert len(modified_values) == 1
 
 
-def test_prov_from_git_error_untracked(tmp_path, datadir, monkeypatch, caplog):
-    """Test that --prov-from-git fails if files are not tracked in git."""
+def test_prov_from_git_skips_untracked(tmp_path, datadir, monkeypatch, caplog):
+    """Test that --prov-from-git skips untracked files with a warning."""
     # Initialize git repo but don't add files
     _run_git(["git", "init"], tmp_path)
     _run_git(["git", "config", "user.email", "test@example.com"], tmp_path)
@@ -236,12 +236,20 @@ def test_prov_from_git_error_untracked(tmp_path, datadir, monkeypatch, caplog):
 
     monkeypatch.chdir(tmp_path)
 
-    with (
-        caplog.at_level(logging.ERROR),
-        pytest.raises(Voc4catError, match="is not tracked in git"),
-    ):
+    contents_before = {f: f.read_bytes() for f in vocdir.rglob("*.ttl")}
+
+    with caplog.at_level(logging.INFO):
         main_cli(["transform", "--prov-from-git", "--inplace", str(vocdir)])
 
+    assert "is not tracked in git" in caplog.text
+    assert "Skipping provenance" in caplog.text
+
+    # Verify that untracked files were not modified (no provenance added)
+    ttl_files = list(vocdir.rglob("*.ttl"))
+    assert len(ttl_files) > 0
+    contents_after = {f: f.read_bytes() for f in ttl_files}
+    assert contents_after == contents_before
+
 
 def test_prov_from_git_error_not_directory(tmp_path, datadir, monkeypatch, caplog):
     """Test that --prov-from-git fails if input is not a directory with .ttl files."""